Credo AI — “The EU AI Omnibus and the Living EU AI Act: A Playbook for Enterprises” (August 2026)
Tag: S-2026-08-25-credo-eu-omnibus-playbook Type: report (vendor playbook / positioning document, ungated; landing page fetched in full — PDF body not read) Author(s): Credo AI Date of source: 2026-08-25 (date listed on credo.ai/resources) Date ingested: 2026-09-08 (14 days after publication — inside the 30-day window, outside the 7-day priority window) Authority weight: low — vendor-authored regulatory summary; useful as a calendar reference but every date must be verified against the Official Journal text before reliance Raw file: /_raw_sources/S-2026-08-25-credo-eu-omnibus-playbook.md
What it claims
Credo AI’s playbook states that the EU AI Omnibus “entered into force on 27 July 2026” and “adjusts specific timelines and mechanisms inside the EU AI Act, which remains fully in force”. Its headline: the largest compliance deadlines moved to 2 December 2027 (stand-alone high-risk systems) and 2 August 2028 (AI built into regulated products); transparency requirements and the Commission’s power to fine model providers went live on 2 August 2026; two new prohibitions follow on 2 December (2026), with penalties up to €35M or 7% of worldwide turnover. Its framing line: “not a single core obligation changed, and new duties still arrived in 2026” — i.e. the delay is a scheduling change, not a substantive relaxation, and enterprises should not pause governance build-out. The resources page also lists Madrigal Pharmaceuticals and Principal (Renee Langeness, Director of Data Governance) as customer references, but these predate the window.
Notable quotes
- “The Omnibus entered into force on 27 July 2026 and adjusts specific timelines and mechanisms inside the EU AI Act, which remains fully in force.” (landing page)
- “The headline change: the largest compliance deadlines moved to 2 December 2027 (stand-alone high-risk systems) and 2 August 2028 (AI built into regulated products).” (landing page)
- “Transparency requirements and the Commission’s power to fine model providers went live on 2 August, and two new prohibitions follow on 2 December, at penalties up to €35M or 7% of worldwide turnover.” (landing page)
- “The headline everyone missed: not a single core obligation changed, and new duties still arrived in 2026.” (landing page)
What’s speculative vs. asserted
Asserted by the vendor: the entry-into-force date (27 July 2026), the 2 Dec 2027 / 2 Aug 2028 deadlines, the 2 Aug 2026 transparency/fining start, “two new prohibitions” on 2 Dec 2026.
Comparison with the vault (not a rewrite of the source): the 2 Dec 2027 / 2 Aug 2028 deadlines and the 2 Aug 2026 transparency/enforcement start match the Commission-sourced record on EU AI Act. The 27 July 2026 entry-into-force date is new to the vault — EU AI Act last recorded the act as signed 8 July 2026 and awaiting Official Journal publication — and is vendor-relayed, unverified. The “two new prohibitions on 2 December” claim is not on EU AI Act and is unverified; it may refer to Omnibus-added Article 5 items but the source does not specify which.
Positioning: the document is a lead-generation asset for Credo’s platform; the “living AI Act” framing supports its continuous-governance product narrative.
Topics this feeds
- Credo AI — regulatory-alignment positioning; first vendor artefact in the vault mapping the post-Omnibus calendar.
- EU AI Act — supplies (unverified) entry-into-force date and the “two new prohibitions” claim for cross-check by the regulatory scan.
- AI Governance Platforms — pure-play regulatory-content positioning after a silent fortnight.
Open questions raised
- Verify the Official Journal publication and entry-into-force date of the Omnibus (27 July 2026 per Credo) — owner: daily regulatory-intelligence scan.
- Which “two new prohibitions” apply from 2 December 2026, and do any touch FS use cases?
- Whether the PDF body (not read) contains any mapping of Omnibus changes to Credo product features.